This proposed project has three (3) specific aims. The first aim is to compare the efficacy of two different psychopharmacologic agents (fluoxetine or divalproex sodium) in the treatment of impulsive aggressive behavior by men and women with Intermittent Explosive Disorder (lED). The second aim is to explore the serotonergic correlates of clinical response to these treatment modalities. The third aim is to compare lED subjects and Normal Control subjects in regard to the assessments of 5-HT receptor function and to the assessment of the 5-HT Transporter. In the context of the first two aims, we will also be testing the possibility of a differential anti-aggressive response to treatment with fluoxetine or divalproex in lED subjects as a function of "severity of lifetime aggressiveness" (e.g., "Moderately Aggressive" vs. "Highly Aggressive"lED subjects). Specifically, we hypothesize that fluoxetine is preferentially efficacious in "moderately aggressive" lED and that divalproex is preferentially efficacious in "highly aggressive" lED. In additIon, this study proposes to more fully examine the nature of the relationship between pretreatment 5-HT receptor function and anti-aggressive response to these agents. Biologic evaluation of the 5-HT system will include m-CPP and ipsapirone Challenges, Platelet Tritiated (3H) Paroxetine 5-HT Transporter Binding and 5-HTRelated DNA Polymorphisms. m-CPP Challenge is included to assess the function of post-synaptic 5-HT-2c receptors (ACTH and Thermal Responses) while Ipsapirone Challenge is included to assess the function of both pre-synaptic 5-HT-1a (Thermal Response)and post-synaptic 5-HT-1a (ACTH Response)receptors. Platelet 5-HTT binding is assessed in this context because fluoxetine (but not divalproex) acts at brain 5- HTT sites and because Platelet (and brain) 5-HTT binding correlates inversely with lifetime aggressiveness. Participants will be randomly assigned, and stratified by lifetime aggressiveness (i.e, Life History of Aggression Score: "Moderate Aggression" = LHA < 17; "High Aggression" = LHA > 18), to one of three (3) treatment conditions as follows: 1) Fluoxetine; 2) Divalproex Sodium; Placebo. Duration of treatment will be twelve weeks.